Ballistic Resistance Testing Techniques
|
|
- Conrad Martin
- 5 years ago
- Views:
Transcription
1 Ballistic Resistance Testing Techniques Thomas Johnson, Laura Freeman, Janice Hester, Jonathan Bell The Problem IDA compared different ballistic test methods under limited sample sizes, typical of conditions in DoD testing. The Department of Defense conducts ballistic resistance testing to estimate the probability that a projectile will perforate the armor of a system under test. Ballistic resistance testing routinely employs sensitivity experiment techniques in which sequential test designs are used to estimate armor performance. Statistical procedures used to estimate the ballistic resistance of armor in the DoD have remained relatively unchanged for decades. New test design methods can lead to improved testing efficiency, which is critical for test and evaluation in the current fiscal climate. In reviewing sequential methods used in DoD and comparing them to those using Monte Carlo simulation, we found that newer test design and analysis techniques provide significant improvements over current methods. Newer methods can reduce test sizes, reduce bias in estimates, and support the estimation of the full probability of perforation curve instead of only a single metric. Armor Testing Various fields use sensitivity experiments to characterize the probability of a binary outcome as a function of a stimulus or stress. DoD ballistic characterization tests employ sensitivity experiments to characterize the probability of perforating armor or another material as a function of a projectile s velocity. One such example is shown in Figure 1. Ballistic characterization tests, which are essential to understanding the vulnerability and lethality of military equipment, are conducted on systems ranging from body armor to vehicle and aircraft armor. In recent years, ballistic characterization tests were conducted for the Enhanced Small Arms Protection Inserts (body armor plates), the Enhanced Combat Helmet, the new floor paneling of the CH-47F, cockpit armor for the KC-46, and armor paneling of the Joint Light Tactical Vehicle, to name a few. Ballistic testing is destructive, and DoD ballistic testing can be expensive in terms of both test and material costs. Sample sizes are generally limited, and ballistic characterization tests are almost always limited to fewer than 2 shots. 12 Research Notes
2 Probability of Perforation Probit model fit 8 percent confidence region V 5 estimate (2413 ft/s) Perforation Non perforation Figure 1. Example Data from a Ballistic Limit Test Ballistic characterization testing frequently focuses on a specific percentile of interest and must therefore provide sufficient data to accurately estimate that percentile. The most common percentile of interest is the velocity at which the projectile has a 5 percent probability of perforating the armor called the ballistic limit or V 5 of the armor for the particular projectile (V 5 Ballistic Test for Armor 1997). Historically, V 5 was sufficient to characterize armor. It can also be estimated more precisely and with fewer shots than other percentiles. With modern armors, however, V 5 might not be sufficient to characterize an armor; users might be more interested in the velocity at which the probability of perforation is 1 percent, or even lower. Estimating V 1 is another reason to explore the efficiency of test techniques based on maximum likelihood estimation (MLE). IDA compared different ballistic test methods under limited sample sizes, typical of conditions in DoD testing. Three simulation studies compared test methods in terms of their efficiency and accuracy at estimating V 5 and V 1. To estimate V 5 and V 1, we used the probit maximum likelihood estimator (Probit-MLE). In a probit model, the response of an armor target to a ballistic projectile can be characterized as perforation or nonperforation. Let y i =1 or denote the binary outcome of the ith shot, perforation or non-perforation, where i =1,2,3,,N are the first, second, third, and final shot, respectively. Let F(x i ) denote the probability that y i =1 for the velocity of the ith shot. The location-scale probit model, which we used in our article to characterize the ballistic resistance of armor, is F(x,μ,σ)=Φ((x-μ)/σ), where Φ is the standard normal cumulative distribution function. We define x P to be the P th quantile of the distribution, where F (x P )=P. In this formulation, μ, the estimator of V 5, is estimated using maximum likelihood estimation, and is 13
3 referred to as the Probit-MLE estimator (Collins 212). Numerous aspects of a ballistic resistance test can affect the quality and consistency of the results: the laboratory setup, location of projectile impact, obliquity angle, temperature, and projectile type, among others. We focused on statistical aspects, namely, the sequential methods. Sequential Methods A sequential method dictates the velocity setting of each projectile fired in a test. We investigated seven sequential methods, chosen based on their prevalence in military armor testing, ease of implementation, and overall effectiveness at estimating V 5 and V 1. The sequential methods compared are: Up and Down Method (UD) (Dixon and Mood 1948) Langlie Method (LM) (Langlie 1962) Delayed Robbins Monroe Method (DRM) (Hodges and Lehmann 1956) Wu s three-phased optimal design approach (3Pod) (Wu and Tian, Three-Phase Sequential Design for Sensitivity Experiments 213) Neyer s Method (NM) (Neyer 1994) Robbins Monroe Joseph Method (RMJ) (Wu and Tian, Three-Phase Optimal Design of Sensitivity Experiments 214) K-in-a-row () (Gezmu 1996). To illustrate their utility, Figure 2 shows notional tests for selected methods. Simulation Comparison Study We compared sequential methods, estimators, and stopping criteria using Monte Carlo simulation. We used a Probit model to represent the true relationship between probability of perforation and projectile velocity. We considered two sets of true parameters that are reflective of the combat helmet example shown in Figure 1: (1) μ = 2,4 ft/s, σ T T =75 ft/s, and (2) μ T =2,4 ft/s, σ T = 15 ft/s. A simulated test is carried out in a manner similar to a physical one except that no projectiles are fired, and the outcome of whether the projectile perforated the armor is determined using a random Bernoulli draw from the probability of perforation estimated from the true model. For example, if a given simulated shot (x) is fired at 2,3 ft/s, according to the first set of true parameters, the probability that that projectile perforates the armor is Φ((xμ T ) ) = Φ((2,3-2,4)/75) =.9, where Φ is the cumulative distribution function from the normal distribution. Then, a random Bernoulli number is generated that has a 9 percent chance of being a perforation. To instill more realism into the simulation, we include a velocity set point error. For each calculated velocity, we add a random error drawn from a uniform distribution between plus or minus 1 ft/s. The simulation employs a full factorial experiment to compare the different test designs in terms of their ability to estimate V 5 and V 1. Table 1 shows the variables considered in the simulation experiment. Ideally, we hope to find a method that results 14 Research Notes
4 Up Down Most commonly used sequential method in the Department of Defense Good for estimating V 5, but not for V 1 If projectile penetrates armor, decrease velocity If projectile does not penetrate armor, increase velocity Langlie Popular method used in the Department of Defense Designed to converge to V 5 Incorporates variable step size in velocity changes Not efficient for estimating V 1 Developed in the early 196s K-in-a-Row If projectile does penetrates armor, decrease velocity If projectile does not penetrate armor k times in a row, increase velocity Useful for estimating V 1 Easy to implement Neyer s Developed by Neyer in 1989 First to propose a systemic method for generating an initial design, which is useful for quickly estimating model coefficients Requires capability to compute maximum likelihood estimation 3Pod Developed by Wu in 213 Involves three phases: 1) Bound the response curve and break separation 2) Place points at the D-optimal location 3) Place points at the percentile of interest Useful for estimating V 5 and V 1 LEGEND V 5 Non Penetration Penetration Fixed Step Size Dynamic Step Sizes k=5 Initial Design D-optimal Points Phase 1 Phase 2 Phase Shot Number Figure 2. Example Tests for Sequential Methods in improved estimates and is robust to poor starting estimates. Note that we intentionally consider cases where the mean and variance of the data are incorrect to represent test cases where there is poor understanding of the armors true performance. The response variables are the median and 15
5 Sequential Method Sample Size N=2, N=4 σt Table 1. Factors and Levels Up and Down (UD), Langlie (LM), Delayed Robbins Monroe (DRM), Three Pod (3Pod), Neyer s (NM), Robbins Monroe Joseph (RMJ), K-in-a-row () 75 ft/s, 15 ft/s σg/σt 1/3,1/2,2,3 µg µt-2σt, µt, µt+2σt interquartile range of the V 5 and V 1 bias. The full factorial experiments consist of 336 trials. One thousand simulations are executed per trial. A simulation is representative of a single live fire test, consisting of either 2 or 4 sequentially fired projectiles. After each simulation, the V 5 bias is calculated as the difference between the assumed true V 5 and the V 5 estimated from the simulation. V 1 is calculated similarly. The median and interquartile range of the V 5 and V 1 bias are the response variables for each factorial trial. Results Figure 3 shows the median and interquartile range of the V 5 and V 1 errors from the 1, simulation runs. The figure illustrates that RMJ and DRM reduce the V 1 median bias more than the other sequential methods. 3Pod is the next best performing, followed by, LM, NM, and finally UD. The advantage of reduced V 1 bias by RMJ comes at the expense of V 5 bias. Figure 3 also shows that the V 1 median error is bias interaction between and σ when G is equal to 2 in one direction for all methods except DRM and RMJ. This result occurs because the other sequential methods place runs closer to V 5, thereby biasing the V 1 estimate closer to V 5. This result is magnified for LM, NM, and UD, since these sequential methods place runs closer to V 5 by design. Figure 4 shows the results of effect screening, which is a more robust way of understanding the results from Figure 3. Effect screening is an efficient way to summarize and compare the results of highly dimensional factorial experiments. The effects show the impacts of the factors, and interactions between factors, on the response variables. The effects are calculated by regressing each simulation outcome (median V 5 and V 1 error) on the factors of the factorial experiment. Coefficient estimates are shown for all main effects and two-factor interactions. The intercept of the regression model is the grand mean of the response variable, shown in the bottom left of the effects plot. The coefficient of a particular level of a factor describes the difference between the grand mean and the average response at that level. Figure 4a shows that the effect that had the largest detrimental effect on the V 1 error was the interaction between and σ when G is equal to 2. In that case, is unable to converge 16 Research Notes
6 UD LM DRM 3POD NM RMJ µ =µ 2σ µ =µ µ =µ + 2 T T G T G T T V 5 Error (ft/s) N = 2 N = 4 1 V 1 Error (ft/s) N = 2 N = 4 1/3 1/ /3 1/ /3 1/2 2 3 Figure 3. Simulation Outputs σ =75 T σ =15 T σ T =75 σ T =15 Runs=2 Runs=4 Runs=2 Runs=4 2σ T µ =µ +2 T T =1/3 =1/2 =2 =3 3POD DRM LM NM RMJ UD σ =75 T σ =15 T σ T =75 σ T =15 Runs=2 Runs=4 Runs=2 Runs=4 2σ T µ =µ +2 T T =1/3 =1/2 =2 =3 3POD DRM LM NM RMJ UD 2σ T +2σ T 2σ T +2σ T (a) Legend =1/3 1 =1/2 =2 1 =3 Coefficient Estimate Intercept =3.9 Units in ft/s 3POD DRM LM NM RMJ UD (b) Legend 4 Coefficient Estimate 4 Intercept =22.8 Units in ft/s =1/3 =1/2 =2 =3 3POD DRM LM NM RMJ UD Figure 4. Screening Results for (a) Magnitude of Median V 1 Error (b) Magnitude of Median V 5 Error 17
7 to V 1 because of its large step size. UD and seem to suffer estimation inaccuracies from interactions, more than the other sequential methods. Meanwhile, 3Pod and NM appear to be most robust. Figure 4b shows that the three best sequential methods for reducing V 5 bias are LM, NM, and 3Pod. RMJ and DRM yield the worst V 5 bias. This result is not surprising because DRM and RMJ forgo initial designs and do not place points near V 5. Meanwhile, 3Pod and NM employ initial designs and D-optimal selection criteria that balance the design space. In general, we found that the methods compared in this study perform commensurate with the goal of the test design. The top three sequential methods that reduce V 1 bias are, in descending order, RMJ, DRM, and 3Pod. However, 3Pod is more robust to incorrectly specified values of μ G and than DRM. We also noted that DRM performs erratically for tests with greater than 2 samples because its step size becomes smaller than the velocity set point error. UD, LM, 3Pod, and NM resulted in the lowest bias on V 5. The 3Pod method appears to be the most robust method of estimating multiple quantiles. Conclusions The DoD uses sensitivity experiments to assess the ballistic resistance of various types of armor. We have shown that employing more recent sensitivity test design methods such as 3Pod and Neyer s Method can lead to improved testing efficiency, increased accuracy, and supports estimation of the entire response curve. Use of these new methods requires that the test community perform real-time statistical analysis of the data during test to select sequential test shots. We have also demonstrated the advantage of using maximum likelihood estimation and generalized linear models in the analysis and execution of ballistic limit testing. Maximum likelihood estimation techniques permit generation of the full perforation response curve, providing more information for the same test resources. Dr. Johnson is a Research Staff Member in IDA s Operational Evaluation Division. He holds a Doctor of Philosophy in aerospace engineering from Old Dominion University. Dr. Freeman is an Assistant Director in IDA s Operational Evaluation Division. She holds a Doctor of Philosophy in statistics from Virginia Polytechnic Institute and State University. 18 Research Notes
8 Dr. Hester is an Adjunct Research Staff Member in IDA s Operational Evaluation Division. She holds a Doctor of Philosophy in physics from Princeton University. Dr. Bell is a Research Staff Member in IDA s Operational Evaluation Division. He holds a Doctor of Philosophy in materials science and engineering from the University of Illinois at Urbana-Champaign. The original article was published in the Institute of Electrical and Electronics Engineers (IEEE) Access, December 214. A Comparison of Ballistic Resistance Testing Techniques in the Department of Defense References Collins, J. C. Quantile response: Practical Sensitivity Testing. Technical Report, Adelphi, MD: US Army Research Lab, 212. Dixon, W. J., and A.M. Mood. A Method for Obtaining and Analyzing Sensitivity Data. (Journal of American Statistisical Association) 43, no. 241 (March 1948): Gezmu, M. The Geometric Up-and-Down Design for Allocating Dosage Levels. Ph.D. Dissertation, Department of Mathematics and Statistics, Washington D.C.: American University, Hodges, J.L., and E.L. Lehmann. Two Approximations to the Robbins-Monroe Process. Proc. Third Berkeley Symp. 1 (1956). Langlie, H.J. A Reliability Test Method for One-Shot Items. Ford Motor Company, Neyer, B. A D-Optimality-Based Sensitivity Test. Technometrics 36, no. 1 (February 1994): V 5 Ballistic Test for Armor. Department of Defense Test Method Standard, Wu, C.F.J., and Y. Tian. Three-Phase Optimal Design of Sensitivity Experiments. Journal of Statistical Planning and Inference 149 (June 214): Wu, C.F.J., and Y. Tian. Three-Phase Sequential Design for Sensitivity Experiments. (Technical report, Georgia Institute of Technology)
Test Strategies for Experiments with a Binary Response and Single Stress Factor Best Practice
Test Strategies for Experiments with a Binary Response and Single Stress Factor Best Practice Authored by: Sarah Burke, PhD Lenny Truett, PhD 15 June 2017 The goal of the STAT COE is to assist in developing
More informationThree-phase Sequential Design for Sensitivity Experiments
Three-phase Sequential Design for Sensitivity Experiments C.F. Jeff Wu Georgia Institute of Technology Sensitivity testing : problem formulation Review of existing procedures for sequential sensitivity
More informationA new procedure for sensitivity testing with two stress factors
A new procedure for sensitivity testing with two stress factors C.F. Jeff Wu Georgia Institute of Technology Sensitivity testing : problem formulation. Review of the 3pod (3-phase optimal design) procedure
More informationA Note on Bayesian Inference After Multiple Imputation
A Note on Bayesian Inference After Multiple Imputation Xiang Zhou and Jerome P. Reiter Abstract This article is aimed at practitioners who plan to use Bayesian inference on multiplyimputed datasets in
More informationStatistical Practice
Statistical Practice A Note on Bayesian Inference After Multiple Imputation Xiang ZHOU and Jerome P. REITER This article is aimed at practitioners who plan to use Bayesian inference on multiply-imputed
More informationWhy Correlation Matters in Cost Estimating
Why Correlation Matters in Cost Estimating Stephen A. Book The Aerospace Corporation P.O. Box 92957 Los Angeles, CA 90009-29597 (310) 336-8655 stephen.a.book@aero.org 32nd Annual DoD Cost Analysis Symposium
More informationDose-Response Analysis Report
Contents Introduction... 1 Step 1 - Treatment Selection... 2 Step 2 - Data Column Selection... 2 Step 3 - Chemical Selection... 2 Step 4 - Rate Verification... 3 Step 5 - Sample Verification... 4 Step
More informationQuantifying Weather Risk Analysis
Quantifying Weather Risk Analysis Now that an index has been selected and calibrated, it can be used to conduct a more thorough risk analysis. The objective of such a risk analysis is to gain a better
More informationQuantile POD for Hit-Miss Data
Quantile POD for Hit-Miss Data Yew-Meng Koh a and William Q. Meeker a a Center for Nondestructive Evaluation, Department of Statistics, Iowa State niversity, Ames, Iowa 50010 Abstract. Probability of detection
More informationGeneral Regression Model
Scott S. Emerson, M.D., Ph.D. Department of Biostatistics, University of Washington, Seattle, WA 98195, USA January 5, 2015 Abstract Regression analysis can be viewed as an extension of two sample statistical
More informationProbability and Probability Distributions. Dr. Mohammed Alahmed
Probability and Probability Distributions 1 Probability and Probability Distributions Usually we want to do more with data than just describing them! We might want to test certain specific inferences about
More informationAssessing Atmospheric Releases of Hazardous Materials
Assessing Atmospheric Releases of Hazardous Materials Nathan Platt and Jeffry Urban The Problem Atmospheric transport and dispersion (AT&D) models play an important role in the Department of Defense because
More informationStatistics Boot Camp. Dr. Stephanie Lane Institute for Defense Analyses DATAWorks 2018
Statistics Boot Camp Dr. Stephanie Lane Institute for Defense Analyses DATAWorks 2018 March 21, 2018 Outline of boot camp Summarizing and simplifying data Point and interval estimation Foundations of statistical
More informationSTAT T&E COE-Report Reliability Test Planning for Mean Time Between Failures. Best Practice. Authored by: Jennifer Kensler, PhD STAT T&E COE
Reliability est Planning for Mean ime Between Failures Best Practice Authored by: Jennifer Kensler, PhD SA &E COE March 21, 2014 he goal of the SA &E COE is to assist in developing rigorous, defensible
More informationCHAPTER 4 VARIABILITY ANALYSES. Chapter 3 introduced the mode, median, and mean as tools for summarizing the
CHAPTER 4 VARIABILITY ANALYSES Chapter 3 introduced the mode, median, and mean as tools for summarizing the information provided in an distribution of data. Measures of central tendency are often useful
More informationEfficient Robbins-Monro Procedure for Binary Data
Efficient Robbins-Monro Procedure for Binary Data V. Roshan Joseph School of Industrial and Systems Engineering Georgia Institute of Technology Atlanta, GA 30332-0205, USA roshan@isye.gatech.edu SUMMARY
More informationResearch Article A Nonparametric Two-Sample Wald Test of Equality of Variances
Advances in Decision Sciences Volume 211, Article ID 74858, 8 pages doi:1.1155/211/74858 Research Article A Nonparametric Two-Sample Wald Test of Equality of Variances David Allingham 1 andj.c.w.rayner
More informationBasics of Uncertainty Analysis
Basics of Uncertainty Analysis Chapter Six Basics of Uncertainty Analysis 6.1 Introduction As shown in Fig. 6.1, analysis models are used to predict the performances or behaviors of a product under design.
More informationThe Union and Intersection for Different Configurations of Two Events Mutually Exclusive vs Independency of Events
Section 1: Introductory Probability Basic Probability Facts Probabilities of Simple Events Overview of Set Language Venn Diagrams Probabilities of Compound Events Choices of Events The Addition Rule Combinations
More informationIntroduction to Statistical Data Analysis Lecture 4: Sampling
Introduction to Statistical Data Analysis Lecture 4: Sampling James V. Lambers Department of Mathematics The University of Southern Mississippi James V. Lambers Statistical Data Analysis 1 / 30 Introduction
More information11. Bootstrap Methods
11. Bootstrap Methods c A. Colin Cameron & Pravin K. Trivedi 2006 These transparencies were prepared in 20043. They can be used as an adjunct to Chapter 11 of our subsequent book Microeconometrics: Methods
More informationLast Lecture. Distinguish Populations from Samples. Knowing different Sampling Techniques. Distinguish Parameters from Statistics
Last Lecture Distinguish Populations from Samples Importance of identifying a population and well chosen sample Knowing different Sampling Techniques Distinguish Parameters from Statistics Knowing different
More information2. Prove that x must be always lie between the smallest and largest data values.
Homework 11 12.5 1. A laterally insulated bar of length 10cm and constant cross-sectional area 1cm 2, of density 10.6gm/cm 3, thermal conductivity 1.04cal/(cm sec C), and specific heat 0.056 cal/(gm C)(this
More informationStat 5101 Lecture Notes
Stat 5101 Lecture Notes Charles J. Geyer Copyright 1998, 1999, 2000, 2001 by Charles J. Geyer May 7, 2001 ii Stat 5101 (Geyer) Course Notes Contents 1 Random Variables and Change of Variables 1 1.1 Random
More informationStatistics - Lecture One. Outline. Charlotte Wickham 1. Basic ideas about estimation
Statistics - Lecture One Charlotte Wickham wickham@stat.berkeley.edu http://www.stat.berkeley.edu/~wickham/ Outline 1. Basic ideas about estimation 2. Method of Moments 3. Maximum Likelihood 4. Confidence
More informationProblems. Suppose both models are fitted to the same data. Show that SS Res, A SS Res, B
Simple Linear Regression 35 Problems 1 Consider a set of data (x i, y i ), i =1, 2,,n, and the following two regression models: y i = β 0 + β 1 x i + ε, (i =1, 2,,n), Model A y i = γ 0 + γ 1 x i + γ 2
More informationForecasting Chapter 3
Forecasting Chapter 3 Introduction Current factors and conditions Past experience in a similar situation 2 Accounting. New product/process cost estimates, profit projections, cash management. Finance.
More informationA simulation study for comparing testing statistics in response-adaptive randomization
RESEARCH ARTICLE Open Access A simulation study for comparing testing statistics in response-adaptive randomization Xuemin Gu 1, J Jack Lee 2* Abstract Background: Response-adaptive randomizations are
More informationA nonparametric two-sample wald test of equality of variances
University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 211 A nonparametric two-sample wald test of equality of variances David
More informationSequential Experimental Designs for Generalized Linear Models
Sequential Experimental Designs for Generalized Linear Models Hovav A. Dror and David M. Steinberg, JASA (2008) Bob A. Salim May 14, 2013 Bob A. Salim Sequential Experimental Designs for Generalized Linear
More informationForecasting. Chapter Copyright 2010 Pearson Education, Inc. Publishing as Prentice Hall
Forecasting Chapter 15 15-1 Chapter Topics Forecasting Components Time Series Methods Forecast Accuracy Time Series Forecasting Using Excel Time Series Forecasting Using QM for Windows Regression Methods
More informationDetermination of cumulative Poisson probabilities for double sampling plan
2017; 3(3): 241-246 ISSN Print: 2394-7500 ISSN Online: 2394-5869 Impact Factor: 5.2 IJAR 2017; 3(3): 241-246 www.allresearchjournal.com Received: 25-01-2017 Accepted: 27-02-2017 G Uma Assistant Professor,
More informationQuantal Response: Practical Sensitivity Testing
Quantal Response: Practical Sensitivity Testing by Joseph C. Collins ARL-TR-6022 June 2012 Approved for public release; distribution is unlimited. NOTICES Disclaimers The findings in this report are not
More informationTRADE SPACE STUDY FOR THE PERFORMANCE OF HARD ARMOR VERSUS THREAT STRIKING VELOCITY AT RANGE
TECHNICAL REPORT NATICK/TR-15/025 AD TRADE SPACE STUDY FOR THE PERFORMANCE OF HARD ARMOR VERSUS THREAT STRIKING VELOCITY AT RANGE by John Kirejczyk August 2015 Final Report September 2013 September 2014
More informationProjectile Motion. Objectives. Materials. Introduction
Projectile Motion Objectives The objectives of this experiment are for you to: 1. Develop confidence in your ability to use the equations of motion to predict the results of an experiment. 2. Gain confidence
More informationParadoxical Results in Multidimensional Item Response Theory
UNC, December 6, 2010 Paradoxical Results in Multidimensional Item Response Theory Giles Hooker and Matthew Finkelman UNC, December 6, 2010 1 / 49 Item Response Theory Educational Testing Traditional model
More informationON REPLICATION IN DESIGN OF EXPERIMENTS
ON REPLICATION IN DESIGN OF EXPERIMENTS Bianca FAGARAS 1), Ingrid KOVACS 1), Anamaria OROS 1), Monica RAFAILA 2), Marina Dana TOPA 1), Manuel HARRANT 2) 1) Technical University of Cluj-Napoca, Str. Baritiu
More information* Tuesday 17 January :30-16:30 (2 hours) Recored on ESSE3 General introduction to the course.
Name of the course Statistical methods and data analysis Audience The course is intended for students of the first or second year of the Graduate School in Materials Engineering. The aim of the course
More informationOptimal Selection of Blocked Two-Level. Fractional Factorial Designs
Applied Mathematical Sciences, Vol. 1, 2007, no. 22, 1069-1082 Optimal Selection of Blocked Two-Level Fractional Factorial Designs Weiming Ke Department of Mathematics and Statistics South Dakota State
More informationBiost 518 Applied Biostatistics II. Purpose of Statistics. First Stage of Scientific Investigation. Further Stages of Scientific Investigation
Biost 58 Applied Biostatistics II Scott S. Emerson, M.D., Ph.D. Professor of Biostatistics University of Washington Lecture 5: Review Purpose of Statistics Statistics is about science (Science in the broadest
More informationChapter 7 Forecasting Demand
Chapter 7 Forecasting Demand Aims of the Chapter After reading this chapter you should be able to do the following: discuss the role of forecasting in inventory management; review different approaches
More informationSeismic Analysis of Structures Prof. T.K. Datta Department of Civil Engineering Indian Institute of Technology, Delhi. Lecture 03 Seismology (Contd.
Seismic Analysis of Structures Prof. T.K. Datta Department of Civil Engineering Indian Institute of Technology, Delhi Lecture 03 Seismology (Contd.) In the previous lecture, we discussed about the earth
More informationComparison of V 50 Shot Placement on Final Outcome
Comparison of V 50 Shot Placement on Final Outcome by R Kinsler and J Collins ARL-RP-0506 November 2014 A reprint from Proceedings of the PASS 2012 Personal Armour Systems Symposium, Nuremberg, Germany,
More information4. STATISTICAL SAMPLING DESIGNS FOR ISM
IRTC Incremental Sampling Methodology February 2012 4. STATISTICAL SAMPLING DESIGNS FOR ISM This section summarizes results of simulation studies used to evaluate the performance of ISM in estimating the
More informationComputing Consecutive-Type Reliabilities Non-Recursively
IEEE TRANSACTIONS ON RELIABILITY, VOL. 52, NO. 3, SEPTEMBER 2003 367 Computing Consecutive-Type Reliabilities Non-Recursively Galit Shmueli Abstract The reliability of consecutive-type systems has been
More informationMaximum Likelihood, Logistic Regression, and Stochastic Gradient Training
Maximum Likelihood, Logistic Regression, and Stochastic Gradient Training Charles Elkan elkan@cs.ucsd.edu January 17, 2013 1 Principle of maximum likelihood Consider a family of probability distributions
More informationOptimal Cusum Control Chart for Censored Reliability Data with Log-logistic Distribution
CMST 21(4) 221-227 (2015) DOI:10.12921/cmst.2015.21.04.006 Optimal Cusum Control Chart for Censored Reliability Data with Log-logistic Distribution B. Sadeghpour Gildeh, M. Taghizadeh Ashkavaey Department
More informationPHYS 1111L - Introductory Physics Laboratory I
PHYS 1111L - Introductory Physics Laboratory I Laboratory Advanced Sheet Projectile Motion Laboratory 1. Objective. The objective of this laboratory is to predict the range of a projectile set in motion
More informationGlossary. Appendix G AAG-SAM APP G
Appendix G Glossary Glossary 159 G.1 This glossary summarizes definitions of the terms related to audit sampling used in this guide. It does not contain definitions of common audit terms. Related terms
More informationContents. Preface to Second Edition Preface to First Edition Abbreviations PART I PRINCIPLES OF STATISTICAL THINKING AND ANALYSIS 1
Contents Preface to Second Edition Preface to First Edition Abbreviations xv xvii xix PART I PRINCIPLES OF STATISTICAL THINKING AND ANALYSIS 1 1 The Role of Statistical Methods in Modern Industry and Services
More informationEPAs New MDL Procedure What it Means, Why it Works, and How to Comply
EPAs New MDL Procedure What it Means, Why it Works, and How to Comply Richard Burrows TestAmerica Inc. 1 A Revision to the Method Detection Limit EPA published a revision to the 40 CFR Part 136 MDL procedure
More informationNumerical simulation of the tumbling of kinetic energy projectiles after impact on ceramic/metal armours
Computational Ballistics II 21 Numerical simulation of the tumbling of kinetic energy projectiles after impact on ceramic/metal armours F. Gálvez 1, S. Chocron 2, D. Cendón 1 & V. Sánchez-Gálvez 1 1 Department
More informationPolynomial Operations Polly s Pasta
Polynomial Operations ACTIVITY 4.2 SUGGESTED LEARNING STRATEGIES: Close Reading, Discussion Group, Create Representations, Think/Pair/Share, Self/Peer Revision and Pizza Supply sells wholesale goods to
More informationPoint and Interval Estimation for Gaussian Distribution, Based on Progressively Type-II Censored Samples
90 IEEE TRANSACTIONS ON RELIABILITY, VOL. 52, NO. 1, MARCH 2003 Point and Interval Estimation for Gaussian Distribution, Based on Progressively Type-II Censored Samples N. Balakrishnan, N. Kannan, C. T.
More informationAIM HIGH SCHOOL. Curriculum Map W. 12 Mile Road Farmington Hills, MI (248)
AIM HIGH SCHOOL Curriculum Map 2923 W. 12 Mile Road Farmington Hills, MI 48334 (248) 702-6922 www.aimhighschool.com COURSE TITLE: Statistics DESCRIPTION OF COURSE: PREREQUISITES: Algebra 2 Students will
More informationInverse Sampling for McNemar s Test
International Journal of Statistics and Probability; Vol. 6, No. 1; January 27 ISSN 1927-7032 E-ISSN 1927-7040 Published by Canadian Center of Science and Education Inverse Sampling for McNemar s Test
More informationIntroduction to Design of Experiments
Introduction to Design of Experiments Jean-Marc Vincent and Arnaud Legrand Laboratory ID-IMAG MESCAL Project Universities of Grenoble {Jean-Marc.Vincent,Arnaud.Legrand}@imag.fr November 20, 2011 J.-M.
More informationTime-varying failure rate for system reliability analysis in large-scale railway risk assessment simulation
Time-varying failure rate for system reliability analysis in large-scale railway risk assessment simulation H. Zhang, E. Cutright & T. Giras Center of Rail Safety-Critical Excellence, University of Virginia,
More informationA MultiGaussian Approach to Assess Block Grade Uncertainty
A MultiGaussian Approach to Assess Block Grade Uncertainty Julián M. Ortiz 1, Oy Leuangthong 2, and Clayton V. Deutsch 2 1 Department of Mining Engineering, University of Chile 2 Department of Civil &
More informationChapter 1 Statistical Inference
Chapter 1 Statistical Inference causal inference To infer causality, you need a randomized experiment (or a huge observational study and lots of outside information). inference to populations Generalizations
More informationObjective Experiments Glossary of Statistical Terms
Objective Experiments Glossary of Statistical Terms This glossary is intended to provide friendly definitions for terms used commonly in engineering and science. It is not intended to be absolutely precise.
More informationEstimation of process parameters to determine the optimum diagnosis interval for control of defective items
Estimation of process parameters to determine the optimum diagnosis interval for control of defective items Abhyuday Mandal Department of Statistics University of Georgia Athens, GA 30602-1952 Joint research
More informationOptimization of Muffler and Silencer
Chapter 5 Optimization of Muffler and Silencer In the earlier chapter though various numerical methods are presented, they are not meant to optimize the performance of muffler/silencer for space constraint
More informationSemiparametric Generalized Linear Models
Semiparametric Generalized Linear Models North American Stata Users Group Meeting Chicago, Illinois Paul Rathouz Department of Health Studies University of Chicago prathouz@uchicago.edu Liping Gao MS Student
More informationReview of Statistics 101
Review of Statistics 101 We review some important themes from the course 1. Introduction Statistics- Set of methods for collecting/analyzing data (the art and science of learning from data). Provides methods
More informationADAPTIVE DESIGNS FOR STOCHASTIC ROOT-FINDING
Statistica Sinica 17(2007), 1549-1565 ADAPTIVE DESIGNS FOR STOCHASTIC ROOT-FINDING V. Roshan Joseph 1, Yubin Tian 2 and C. F. Jeff Wu 1 1 Georgia Institute of Technology and 2 Beijing Institute of Technology
More informationP E R S P E C T I V E S
PHOENIX CENTER FOR ADVANCED LEGAL & ECONOMIC PUBLIC POLICY STUDIES Econometric Analysis of Broadband Subscriptions: A Note on Specification George S. Ford, PhD May 12, 2009 Broadband subscriptions are
More informationSmall Sample Corrections for LTS and MCD
myjournal manuscript No. (will be inserted by the editor) Small Sample Corrections for LTS and MCD G. Pison, S. Van Aelst, and G. Willems Department of Mathematics and Computer Science, Universitaire Instelling
More informationA process capability index for discrete processes
Journal of Statistical Computation and Simulation Vol. 75, No. 3, March 2005, 175 187 A process capability index for discrete processes MICHAEL PERAKIS and EVDOKIA XEKALAKI* Department of Statistics, Athens
More informationa table or a graph or an equation.
Topic (8) POPULATION DISTRIBUTIONS 8-1 So far: Topic (8) POPULATION DISTRIBUTIONS We ve seen some ways to summarize a set of data, including numerical summaries. We ve heard a little about how to sample
More informationSummarizing Measured Data
Summarizing Measured Data 12-1 Overview Basic Probability and Statistics Concepts: CDF, PDF, PMF, Mean, Variance, CoV, Normal Distribution Summarizing Data by a Single Number: Mean, Median, and Mode, Arithmetic,
More informationTesting the Limits of Latent Class Analysis. Ingrid Carlson Wurpts
Testing the Limits of Latent Class Analysis by Ingrid Carlson Wurpts A Thesis Presented in Partial Fulfillment of the Requirements for the Degree Master of Arts Approved April 2012 by the Graduate Supervisory
More informationAdditional Problems Additional Problem 1 Like the http://www.stat.umn.edu/geyer/5102/examp/rlike.html#lmax example of maximum likelihood done by computer except instead of the gamma shape model, we will
More informationGLOSSARY. a n + n. a n 1 b + + n. a n r b r + + n C 1. C r. C n
GLOSSARY A absolute cell referencing A spreadsheet feature that blocks automatic adjustment of cell references when formulas are moved or copied. References preceded by a dollar sign $A$1, for example
More informationFall 2013 Northwestern University STAT Section 23 Introduction to Statistics. Final Exam. Name:
Fall 2013 Northwestern University STAT 202-0 Section 23 Introduction to Statistics Final Exam Name: December 13, 2013 1. Please do not leave blank for any question. 2. There are 15 questions, each question
More informationParameter Estimation for Random Differential Equation Models
Parameter Estimation for Random Differential Equation Models H.T. Banks and Michele L. Joyner Center for Research in Scientific Computation North Carolina State University Raleigh, NC, United States and
More informationBootstrapping the Confidence Intervals of R 2 MAD for Samples from Contaminated Standard Logistic Distribution
Pertanika J. Sci. & Technol. 18 (1): 209 221 (2010) ISSN: 0128-7680 Universiti Putra Malaysia Press Bootstrapping the Confidence Intervals of R 2 MAD for Samples from Contaminated Standard Logistic Distribution
More informationStatistics 3 WEDNESDAY 21 MAY 2008
ADVANCED GCE 4768/01 MATHEMATICS (MEI) Statistics 3 WEDNESDAY 1 MAY 008 Additional materials: Answer Booklet (8 pages) Graph paper MEI Examination Formulae and Tables (MF) Afternoon Time: 1 hour 30 minutes
More informationA COMPARISON OF POISSON AND BINOMIAL EMPIRICAL LIKELIHOOD Mai Zhou and Hui Fang University of Kentucky
A COMPARISON OF POISSON AND BINOMIAL EMPIRICAL LIKELIHOOD Mai Zhou and Hui Fang University of Kentucky Empirical likelihood with right censored data were studied by Thomas and Grunkmier (1975), Li (1995),
More informationComparison of Least Square Estimators with Rank Regression Estimators of Weibull Distribution - A Simulation Study
ISSN 168-80 Journal of Statistics Volume 0, 01. pp. 1-10 Comparison of Least Square Estimators with Rank Regression Estimators of Weibull Distribution - A Simulation Study Abstract Chandika Rama Mohan
More informationExtreme Rain all Frequency Analysis for Louisiana
78 TRANSPORTATION RESEARCH RECORD 1420 Extreme Rain all Frequency Analysis for Louisiana BABAK NAGHAVI AND FANG XIN Yu A comparative study of five popular frequency distributions and three parameter estimation
More informationPILOT OF AN EXTERNAL QUALITY ASSURANCE PROGRAMME FOR CRYSTALLINE SILICA
PILOT OF AN EXTERNAL QUALITY ASSURANCE PROGRAMME FOR CRYSTALLINE SILICA MJ Shai, National Institute for Occupational Health ABSTRACT The mining industry has committed itself to reducing work-related silicosis.
More informationEPSY 905: Fundamentals of Multivariate Modeling Online Lecture #7
Introduction to Generalized Univariate Models: Models for Binary Outcomes EPSY 905: Fundamentals of Multivariate Modeling Online Lecture #7 EPSY 905: Intro to Generalized In This Lecture A short review
More informationSwarthmore Honors Exam 2012: Statistics
Swarthmore Honors Exam 2012: Statistics 1 Swarthmore Honors Exam 2012: Statistics John W. Emerson, Yale University NAME: Instructions: This is a closed-book three-hour exam having six questions. You may
More informationBy choosing to view this document, you agree to all provisions of the copyright laws protecting it.
Copyright 2017 IEEE. Reprinted, with permission, from Sharon L. Honecker and Umur Yenal, Quantifying the Effect of a Potential Corrective Action on Product Life, 2017 Reliability and Maintainability Symposium,
More informationProbability Methods in Civil Engineering Prof. Dr. Rajib Maity Department of Civil Engineering Indian Institution of Technology, Kharagpur
Probability Methods in Civil Engineering Prof. Dr. Rajib Maity Department of Civil Engineering Indian Institution of Technology, Kharagpur Lecture No. # 36 Sampling Distribution and Parameter Estimation
More informationSection 3 : Permutation Inference
Section 3 : Permutation Inference Fall 2014 1/39 Introduction Throughout this slides we will focus only on randomized experiments, i.e the treatment is assigned at random We will follow the notation of
More information1 The Classic Bivariate Least Squares Model
Review of Bivariate Linear Regression Contents 1 The Classic Bivariate Least Squares Model 1 1.1 The Setup............................... 1 1.2 An Example Predicting Kids IQ................. 1 2 Evaluating
More informationUNCERTAINTY ANALYSIS IN BURIED LANDMINE BLAST CHARACTERIZATION DRDC-RDDC-2016-N029
UNCERTAINTY ANALYSIS IN BURIED LANDMINE BLAST CHARACTERIZATION DRDC-RDDC-2016-N029 M. Ceh, T. Josey, W. Roberts Defence Research and Development Canada, Suffield Research Centre, PO Box 4000, Stn Main,
More informationVerification & Validation of Physics- Based Models for Blast Applications
Verification & Validation of Physics- Based Models for Blast Applications Amy Tank and Dr. James Walbert SURVICE Engineering Physics-Based Modeling in Design & Development for U.S. Defense November 5-8,
More informationTwo hours. To be supplied by the Examinations Office: Mathematical Formula Tables and Statistical Tables THE UNIVERSITY OF MANCHESTER.
Two hours MATH38181 To be supplied by the Examinations Office: Mathematical Formula Tables and Statistical Tables THE UNIVERSITY OF MANCHESTER EXTREME VALUES AND FINANCIAL RISK Examiner: Answer any FOUR
More informationTrip Generation Characteristics of Super Convenience Market Gasoline Pump Stores
Trip Generation Characteristics of Super Convenience Market Gasoline Pump Stores This article presents the findings of a study that investigated trip generation characteristics of a particular chain of
More informationRecursive Estimation
Recursive Estimation Raffaello D Andrea Spring 08 Problem Set : Bayes Theorem and Bayesian Tracking Last updated: March, 08 Notes: Notation: Unless otherwise noted, x, y, and z denote random variables,
More informationTime Truncated Modified Chain Sampling Plan for Selected Distributions
International Journal of Research in Engineering and Science (IJRES) ISSN (Online): 2320-9364, ISSN (Print): 2320-9356 Volume 3 Issue 3 ǁ March. 2015 ǁ PP.01-18 Time Truncated Modified Chain Sampling Plan
More informationDirectionally Sensitive Multivariate Statistical Process Control Methods
Directionally Sensitive Multivariate Statistical Process Control Methods Ronald D. Fricker, Jr. Naval Postgraduate School October 5, 2005 Abstract In this paper we develop two directionally sensitive statistical
More informationStatistics Diagnostic. August 30, 2013 NAME:
Statistics Diagnostic August 30, 013 NAME: Work on all six problems. Write clearly and state any assumptions you make. Show what you know partial credit is generously given. 1 Problem #1 Consider the following
More informationApplications of Reliability Demonstration Test
Applications of Reliability Demonstration Test Winson Taam Applied Statistics, NST, BR&T Jun 3, 2009 BOEING is a trademark of Boeing Management Company. EOT_RT_Sub_Template.ppt 1/6/2009 1 Outline Concept
More informationGlossary for the Triola Statistics Series
Glossary for the Triola Statistics Series Absolute deviation The measure of variation equal to the sum of the deviations of each value from the mean, divided by the number of values Acceptance sampling
More informationIndustrial Engineering Prof. Inderdeep Singh Department of Mechanical & Industrial Engineering Indian Institute of Technology, Roorkee
Industrial Engineering Prof. Inderdeep Singh Department of Mechanical & Industrial Engineering Indian Institute of Technology, Roorkee Module - 04 Lecture - 05 Sales Forecasting - II A very warm welcome
More informationSupplemental Information Likelihood-based inference in isolation-by-distance models using the spatial distribution of low-frequency alleles
Supplemental Information Likelihood-based inference in isolation-by-distance models using the spatial distribution of low-frequency alleles John Novembre and Montgomery Slatkin Supplementary Methods To
More information